I. Field of the Invention
The invention generally relates to dense wavelength division multiplexers (DWDM) and in particular to a technique for locking transmission wavelengths of individual lasers of the DWDM.
II. Description of the Related Art
A DWDM is a device for simultaneously transmitting a set of discrete. information channels over a single fiber optic transmission line. A conventional fiber optic transmission line is capable of reliably transmitting signals within a bandwidth of 1280 to 1625 nanometers (nm), the “low loss” region for silica fiber. Within that overall bandwidth, the International Telecommunications Union (ITU) has defined various transmission bands and specified certain transmission channel protocols for use within each transmission band. One example of a transmission band is the ITU “C” band, which extends 40 nm from 1525 nm to 1565 nm. Within the C band, specific transmission channel protocols of 40, 80, or 160 discrete channels are defined and, for each protocol, the ITU has defined a grid of transmission wavelengths, with each line corresponding to an acceptable transmission wavelength. The protocols have been defined to ensure that all DWDM transmission and reception equipment are fabricated to operate at the same wavelengths. For the 40-channel protocol, the corresponding ITU grid has 40 lines with channel spacing of 0.8 nm; for the 80-channel protocol, the corresponding ITU grid has 80 lines with channel spacing of 0.4 nm; and so forth. Additional protocols have been proposed, including 320 channel and 640 channel protocols. Maximum theoretical transmission frequencies for the various ITU protocols are as follows: 100 GHz for the 40 channel protocol; 50 GHz for the 80 channel protocol; 25 GHz for the 160 channel protocol; 12.5 GHz for the 320 channel protocol; and 6.25 GHz for the 640 channel protocol. Closer channel spacing necessitates a lower modulation rate since channel spacing must be larger than the modulation frequency high frequency modulation requires suitable optic fibers, as well as appropriate transmission and receiving equipment. Current state-of-the-art DWDMs typically employ a 40 channel ITU protocol but transmit at 2.5 GHz, well below the theoretical maximum. Other exemplary ITU transmission bands are the S- and L-bands.
To simultaneously transmit the set of channels on a fiber optic cable, a conventional DWDM employs a set of the individual distributed feedback (DFB) lasers, with one DFB laser per channel and with the DFB configured to transmit. FIG. 1 illustrates a DWDM 100 having forty individual DFB lasers 102 for transmitting optical signals via a single optic fiber 104. An optical multiplexer 106 couples signals received from the individual DFBs via a set of intermediate optic fibers 107 into output optic fiber 104. Each DFB laser transmits at a different wavelength of the 40-channel ITU C band. This enables forty separate channels of information to be transmitted via the single optical fiber 104 to a de-multiplexer (not shown) provided at the far end of the optical fiber.
To permit the DWDM to transmit forty separate channels simultaneously, each individual DFB must be tuned to a single ITU transmission channel wavelength. A DFB laser can be tuned only within a narrow wavelength band, typically about 2 nm in width. Hence, for the 40-channel protocol of the ITU, C band having 0.8 nm transmission line spacing, the typical DFB can only be tuned to one of a few adjacent lines out of the total of 40 lines of the ITU grid. Traditionally each individual DFB laser is manually calibrated at the factory to emit at a corresponding one of the ITU transmission lines. This is achieved by adjusting the laser operating temperature and current to obtain the desired wavelength.
The laser is then, in some implementations, locked to the target wavelength by routing the output beam from each DFB laser through a corresponding manually tunable etalon. (The etalons are not shown in FIG. 1.) A manually tunable etalon is an optical device that produces a periodically-varying transmission spectrum as a function of laser wavelength. By tilting the etalon relative to the DFB laser beam path, a transmission peak of the etalon can be made coincident with the target ITU channel. The wavelength of an etalon transmission peak is calibrated to one of the ITU transmission lines by manually adjusting the angle of the etalon while monitoring the wavelength output from the etalon using an optical wavelength analyzer. The angle of the etalon is adjusted until the output wavelength is properly aligned with one of the ITU transmission lines, then the etalon is mounted in place in an attempt to lock the output wavelength of etalon to the selected ITU transmission line. This is a difficult and time-consuming process requiring skilled technicians resulting in the alignment of the etalon transmission peaks with several ITU transmission lines. In addition, if transmission over a certain range of ITU transmission lines is desired, multiple tunable lasers are required to ensure adequate coverage. Such an arrangement is costly and does not provide rapid switching between ITU transmission lines. Furthermore, mechanical or thermal drift of the etalon over time often moves the transmission peak away from the target ITU channel, which requires recalibration.
Once the DFB lasers of a single DWDM are properly aligned with the ITU grid, the DWDM may then be used for transmitting signals over a fiber optic line, such as for transmitting digital data over computer networks or for transmitting television signals from a television network to one of its affiliates. A single DWDM must be provided for use with each fiber optic line employed for DWDM transmissions and hence a single customer installation, such as a television broadcast center, may require large numbers of DWDMs. If one of the DFB lasers within a DWDM drifts from its corresponding ITU transmission line or otherwise malfunctions, the entire DWDM typically needs to be replaced to permit the malfunctioning DWDM to be returned to the factory to be re-calibrated or otherwise fixed. As a result, the cost of maintaining a set of DWDMs is often substantial. To help remedy this problem, some DWDMs are provided with an additional widely tunable laser (WTL), which can be tuned separately to any one of the ITU grid lines. Hence, if one of the DFB lasers malfunctions, the single WTL can be tuned to the corresponding transmission wavelength of the DFB to thereby permit the DWDM to continue to operate. Additional WTLs can be supplied with a DWDM to accommodate the failure of two or more DFB channels, and such “sparing” is a major advantage a WTL over a DFB. However, conventional WTLs cannot simply and accurately be tuned to any target ITU channel at a customer installation and must be calibrated at the factory for operation at a specific channel.
Another problem associated with employing DFB lasers within DWDMs is that, because each DFB laser can only be tuned within a narrow range of about 2 nm, each DFB laser can only be calibrated to one of a few adjacent ITU transmission wavelength lines. It is sometimes desirable to configure the DWDM to use many lasers for transmitting at a single ITU transmission line to provide more bandwidth on that channel. When using DFB lasers, no more than two or three of the lasers can be calibrated to a single ITU transmission line. Hence, in some DWDMs, WTLs are used exclusively instead of DFB lasers, thus permitting any of the lasers to be manually calibrated at the customers installation to transmit on any of the ITU transmission lines. Although the use of WTLs remedies many of the problems associated with using DFB lasers, WTLs are difficult and expensive to fabricate and initially calibrate, and are susceptible to wavelength drift requiring frequent recalibration at the customers installation by trained technicians and hence necessitating high overall installation and maintenance costs.
Thus, whether using DFB lasers or WTLs within a DWDM, significant problems arise in achieving and maintaining proper wavelength calibration of the lasers to permit reliable operation of the DWDM. Accordingly, there was a need to provide an efficient method and system for calibrating transmission lasers within a DWDM and it was to that end that the invention of the senior parent applications were primarily directed. Briefly, the senior parent patent applications involve, inter alia, techniques for calibrating a transmission WTL of a DWDM using an etalon and a gas cell having acetylene, hydrogen cyanide or carbon dioxide. Initially, the absolute transmission wavelengths of the WTL are calibrated by routing an output beam from the WTL through the etalon and through the gas cell while varying tuning parameters of the WTL to thereby generate an etalon spectrum and a gas absorption spectrum both as functions of the tuning parameters. The etalon and gas absorption spectra are compared, along with input reference information specifying gas absorption as a function of absolute wavelength, to determine the absolute transmission wavelength for the WTL as a function of the tuning parameters. The WTL is then tuned to align the transmission wavelength of the WTL to an ITU transmission grid line. By tuning the output wavelength of the WTL using an etalon in combination with a gas absorption cell, the WTL can be quickly, easily and precisely set to a selected ITU transmission grid line at a customer installation. The tuning process can be periodically repeated to maintain precise tuning of the WTL despite possible temperature or mechanical drift of the various components. In one implementation, a wavelength mapper is provided for manually connecting to a WTL to tune the WTL to a selected ITU transmission gridline. In another implementation, the wavelength mapper is permanently attached to the WTL along with a wavelength locker to lock the WTL to an ITU transmission gridline.
Insofar as wavelength locking is concerned, the parent applications describe a wavelength locker employing a temperature-controlled etalon. After the aforementioned wavelength mapping steps are performed to determine the absolute wavelength of the laser as a function of the laser tuning parameters, tuning parameters are applied to the laser to tune the laser to a selected transmission wavelength, such as an ITU channel wavelength. A temperature offset is applied to the etalon of the wavelength locker to vary the wavelengths of the transmission peaks of the etalon until one of the transmission peaks is precisely aligned with the selected wavelength. Any drift of the laser from the etalon transmission peak is detected and the tuning parameters applied to the laser are automatically adjusted to compensate for the drift. The temperature of the etalon is precisely maintained so that the etalon transmission peak does not drift from the selected wavelength. In this manner, the main output beam of the laser remains locked on the absolute wavelength of the selected transmission channel despite possible variations in the output characteristics of the laser. Periodically, the system can be recalibrated using the known absolute wavelengths of the gas absorption chamber to ensure that the transmission peak of the etalon has not drifted from the absolute wavelength of the selected transmission channel.
Although the parent applications describe highly useful techniques for mapping the transmission wavelengths of lasers within a DWDM and for locking the transmission wavelengths to ITU grid lines, room for further improvement remains, particularly insofar as the design and fabrication of the wavelength locker is concerned. For practical applications, the wavelength locker should be highly miniaturized and configured so as to consume relatively little power. The wavelength locker also should be sufficiently durable to operate reliably over a ten- or twenty-year lifetime. Ideally, the wavelength locker should be designed so as to work in combination with any of a wide variety of ITU transmission protocols and fiber optic transmission rates, both existing and proposed and to allow rapid switching between ITU grid lines. Perhaps most importantly, the wavelength locker should be designed so as to be sufficiently inexpensive for practical use. Difficulties arise in each of these areas.
Each WTL for use in a DWDM is typically provided in a miniature “butterfly” package for mounting to a circuit board also containing microcontrollers and other components. The circuit boards are mounted in a parallel array within the DWDM with, typically, one board per ITU channel. Hence, a forty ITU channel DWDM employs forty circuit boards; an eighty ITU channel DWDM employs eighty circuit boards. Current state-of-the-art WTLs typically draw about ten watts of power, thus requiring 400 watts of power or more for the a forty channel DWDM and correspondingly more power for 80 or 160 channel DWDMs. A significant portion of the power is consumed by thermoelectric (TE) coolers provided for controlling the temperature of the semiconductor laser of the WTL. With the WTLs already consuming considerable power, it is particularly important that the wavelength locker be configured so as to minimize power consumption, particularly the temperature-controlled etalon. Minimizing power consumption, however, typically requires that the etalon be configured to provide numerous closely-spaced transmission peaks (i.e. to have a narrow free spectral range) such that relatively little heating or cooling is required to expand or contract the etalon or change its index of refraction sufficiently enough to align one of the transmission lines of the etalon with a selected ITU grid line. Using numerous closely spaced peaks, however, increases the risk that the wavelength locker will lock the transmission wavelength of the WTL to the wrong wavelength. Also, to provide numerous closely-spaced transmission peaks, the etalon typically must be configured to have a very short optical axis, thereby making it more difficult to fabricate and align.
Moreover, difficulties arise in adequately insulating the temperature-controlled etalon so as to minimize power loss and to ensure a minimal temperature gradient within the etalon. Any significant temperature gradient within the etalon tends to degrade the finesse of the etalon (i.e. the sharpness of the individual etalon lines) thus making it difficult to achieve precise wavelength locking. Likewise, any slight misalignment of the etalon or any slight imprecision in reflection coatings of the etalon reduces the degree of finesses. Lack of adequate insulation, of course, also increases power consumption and generates a greater amount of waste heat, which may affect the ability of the TE cooler of the laser to efficiently control the temperature of the laser, particularly if the etalon is mounted closely adjacent to the laser within the butterfly package. Typically, manufacturing protocols for DWDMs specify that the DWDM must operate at 70 degrees Celsius or less, thus putting further limitations on the design of the temperature-controlled etalon. It is difficult, therefore, to provide a temperature-controlled etalon and other wavelength locker, which achieves the requisite degree of finesse for precise wavelength locking while also minimizing power consumption, even for use with just one ITU channel protocol. Ideally, however, the temperature-controlled etalon and other components of the wavelength locker should be configured to work with any of a variety of ITU channel protocols, such as 40 to 640 channels, and with any of a variety of transmission frequencies, such as from 2.5 GHz to 100 GHz. Also, ideally, the wavelength locker is sufficiently miniaturized to mount inside the butterfly package of the WTL to minimize overall circuit board space.
For all of the foregoing reasons, it was desirable to provide improved methods and systems for implementing a wavelength locker for use in locking the transmission wavelength of a laser of a DWDM, which is highly miniaturized, achieves low implementation costs and operating costs, consumes relatively little power, works in combination with any of a wide variety of ITU transmission protocols, rapidly switches between ITU grid lines, and is sufficiency durable to reliably operate for ten to twenty years.
The invention first described in the parent application filed Jan. 31, 2001 was directed to providing just such an improved wavelength locker. Briefly, to lock a WTL to an ITU grid line, a portion of the output beam from the WTL is routed through the etalon to split the beam into a set of transmission lines for detection by a detector. Another portion of the beam is routed directly to a laser wavelength detector via a beam splitter interposed between the laser and the etalon. A wavelength-locking controller compares the signals from the two detectors and adjusts the temperature of the etalon to align the wavelength of one of the transmission lines of the etalon with the wavelength of the output beam. In this arrangement, the wavelength-locking controller operates to control the WTL in a feedback loop to lock the laser to the etalon line. The wavelength-locking controller thereafter monitors the temperature of the etalon and keeps the temperature constant to prevent any wavelength drift in the etalon. In one specific example, the etalon is a silicon etalon configured to have finesse of about 20 and to provide a free spectral range of about 8 GHz With these parameters, the system is able to lock the wavelength of the WTL to within a precision of about 0.2 GHz. In another example, the etalon is first calibrated during manufacture to determine a “set point” operating temperature sufficient to align transmission peaks of the etalon with desired ITU grid lines. The etalon is thereafter mounted within a WTL and the etalon is adjusted to the set point temperature so as to align transmission peaks of the etalon with the desired ITU grid lines to permit wavelength locking. This later technique allows for rapid switching between channels and obviates the need for a gas cell within the WTL.
Still further improvements are achievable. In particular, with the arrangement just summarized, a beam splitter is employed to split the laser beam into two beams—one to the laser wavelength detector and another to the etalon. The beam splitter causes amplitude variations in the two beams, which are out of phase from one another making it more difficult to lock the laser to one of the ITU channels. Accordingly, it was desirable to provide an alternative configuration permitting easier wavelength locking. The invention first described in the parent application filed Sep. 28, 2001 was partly directed to providing just such a configuration. Briefly, a wavelength locker was provided wherein the laser wavelength detector receives a portion of the laser beam (to detect the power of the laser beam) directly from the laser so that phase characteristics of the laser beam are not affected by an intervening beamsplitter thereby permitting improved wavelength locking. In one example, the laser, the laser wavelength detector, the etalon and the etalon wavelength detector are all positioned along a common axis, with the laser wavelength detector interposed between the laser and the etalon. The laser wavelength detector is offset slightly from the common axis to permit most of the laser beam to pass directly to the etalon while capturing a smaller portion of the beam for use in detecting the wavelength of the laser. The laser wavelength detector preferably captures 30% of the beam of the laser, permitting the etalon to capture the remaining 70%, and thereby achieving a lock point ratio of about 1.0 for use in side locking. By positioning the laser wavelength detector along the initial linear portion of the optical path of the laser, the laser wavelength detector directly receives a portion of the output beam from the laser and hence the amplitude of the beam as a function of wavelength is unaffected, as might otherwise occur with the detector positioned off-axis and a beam splitter employed to reflect a portion of the laser beam to the detector.
Another concern lies in the possibility of optical frequency chirp affecting signals transmitted along the optic fiber at one of the ITU gridline wavelengths. Briefly, optical chirp occurs when a current source used to modulate the transmission laser causes dynamic changes in the index of refraction of a laser junction or other optical transmitter. A dynamic change in the index of refraction in turn causes a dynamic change in the actual transmission frequency of an optical pulse transmitted into the optic fiber. As a result, the time average of the optical transmission may indicate that a leading edge of an optical pulse has a slightly different frequency than the trailing edge of the pulse. Although the initial frequency differential between the leading and trailing edges of the pulse may be slight, chromatic dispersion inherent in optic fibers causes the leading and trailing edges to propagate at different speeds resulting in potentially significant distortion of the optical pulse, particularly over long haul optic fiber transmission systems. The distortion limits either the maximum frequency of signal transmission modulation or the maximum distance at which signals can be reliably transmitted.
Accordingly, it was desirable to provide a system for limiting optical frequency chirp, particularly for use within system transmitting on ITU channel wavelengths and other aspects of the invention described in the parent application filed Sep. 28, 2001 were directed to providing just such a system. In one example, the system includes a laser controlled to transmit at a frequency selected from a group of channel frequencies having fixed wavelength spacing. An etalon is mounted between the laser and an output optic fiber. The etalon has a free spectral range (FSR) equal to the fixed wavelength spacing and is tuned so that fringes produced by the etalon are aligned with the transmission channels. Preferably, the etalon has a finesse set such that the width of the fringe is approximately 1 GHz. By aligning the fringes of the etalon with the transmission channel frequencies, only those portions of signals transmitted at frequencies aligned with one of the selected transmission frequencies are passed by the etalon. Portions of signals with frequencies offset from the selected transmission frequencies are filtered out, including any portions of signals having a frequency offset as a result of transmission signal chirp. In this manner, chirp is reduced or completely eliminated by the etalon regardless of which of the selected frequencies is used for transmitting the signals. With chirp eliminated, the signals can be reliably transmitted over longer fiber optic cables with higher modulation rates.
Still further room for improvement remains, particularly insofar as configuring the wavelength locker such that phase characteristics of the laser beam are not affected by any intervening beam splitters so as to permit improved wavelength locking. As noted, in the technique of the parent application filed Sep. 28, 2001, the laser wavelength detector is offset slightly from the common axis of the wavelength locker to permit most of the laser beam to pass directly to the etalon while capturing a smaller portion of the beam (to detect the power of the laser beam) for use in detecting the wavelength of the laser. Although the off-set wavelength locker is effective in eliminating the need for beam splitters, certain design and fabrication issues arise as a result of the need to offset the laser wavelength detector relative to the other optical components. In the example described, both the laser wavelength detector and etalon detector are the same size. The laser wavelength detector is preferably offset from the center of the laser beam so as to capture only 30% of the input beam while permitting the remaining 70% to pass on to the etalon so as to achieve a lock point ratio of about 1.0. In other words, the laser wavelength detector is positioned so as to extend into the laser beam just enough to occlude 30% of the beam. This can be achieved by first determining the amount of offset required to occlude 30% of the laser beam based on the diameter of the detector and on the intensity profile of the laser beam. However, if the laser beam is either non-collimated or collimated using only a relatively inexpensive ball lens, the intensity profile can be relatively imprecise, such that an accurate offset cannot be easily determined in advance. The wavelength locker can be designed to incorporate a more precise collimator, such as an aspheric lens or a graded index (GRIN) lens, but such collimators are relatively expensive. Moreover, even if the correct offset can be calculated in advance, manufacturing tolerances may be such that it is difficult to achieve the offset during automated fabrication. In this regard, a small variation in the offset of the laser wavelength detector can result in a fairly significant variation in the relative beam percentages captured by the two detectors. The laser wavelength detector can be manually adjusted during fabrication until it occludes 30% of the laser beam, but manual adjustment is relatively expensive and time-consuming. Although, achieving a precise lock point ratio of 1.0 is not critical, it is certainly preferred and, accordingly, it would be desirable to provide an alternative configuration of the wavelength locker which permits the lock point ratio of 1.0 to be more readily achieved and it is to this end that aspects of the invention of the present Continuation-In-Part application are directed.
Also, there is room for further improvement in the manner by which the components of the wavelength locker are mounted. In the examples described in the various parent applications, epoxy is typically used to mount the various optical components to a base or substrate. However, concerns arise with respect to the reliability is of epoxy, and particularly with regard to the possibility of out-gassing of material from the epoxy with time, which could coat or otherwise damage the optical components. Accordingly, it would also be desirable to provide an improved mounting arrangement, which does not employ epoxy but which nevertheless provides easy and reliable mounting. It is to this end that other aspects of the invention of the present Continuation-In-Part application are directed.